Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only declare write/non-idempotent/non-destructive/open-world traits; the description adds the substantive behavioral detail: creation produces a mutable draft ('the only mutable surface'), name-only yields an empty draft, and results may carry llmContext guidance that should be followed. All claims are consistent with annotations — readOnlyHint=false aligns with a create operation, so there is no contradiction.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is dense but tightly packed: the core purpose is front-loaded in the first clause, and every subsequent clause — init modes, mutual exclusivity, lifecycle, llmContext note — carries distinct operational value. There is no filler, no restatement of the title, and no repetition of schema content beyond the single cloneSplitterId overlap.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For an open-world creation tool with an output schema to cover return values, the description provides everything an agent needs: the three initialization paths, the decision rule between them, the full lifecycle of the created artifact, and the llmContext operational warning. The environment/workspaceId constraints are fully handled by the schema's own descriptions, so nothing critical is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With 100% schema coverage the baseline is 3, but the description adds real value: the mutual-exclusivity constraint between config and cloneSplitterId is absent from the schema, the pointer to get_documentation for hand-authoring configs gives construction guidance, and 'name alone creates an empty draft' ties the name parameter to a default behavior. Minor redundancy with cloneSplitterId's existing schema description keeps it one notch below perfect.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with 'Create a saved, reusable splitter (split group)', a specific verb+resource statement that clarifies both the action and what a splitter is (a saved, reusable split group). It also establishes the crucial draft-versus-published distinction, which differentiates this tool from siblings like publish_splitter_version and other create_* tools.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives explicit init-mode guidance: 'Start from config ... or cloneSplitterId ... — mutually exclusive; name alone creates an empty draft' tells the agent exactly how to choose between the three input paths. It then names the exact alternatives for the follow-on workflow: 'edit it with update_splitter, freeze it with publish_splitter_version, run it with split_document.'

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.